For a long time I've mulled over the idea of getting some kind of model helicopter rig and fixing a good P&S camera to it to give myself, and the people I'm hoping to pass my work down to, more interesting views of the places I go to. A couple of years ago I was looking at $500 helicopter and undercarriages for cameras that were roughly as expensive. I never got serious about it, but I dreamed, anyway.
Technology's really catching up. Now for about $300 you get pretty much the whole shebang. There's something called the AR Drone 2.0 that's essentially nothing but an HD video camera with four rotors. Check this out.
Pretty impressive, given that it's all for about the price of an average mid-level P&S camera alone nowadays, and considerably less than the cost of a high-end one.
There are things I'm iffy about with the design. While I get the idea and some of the advantages of the paradigm, I am not pleased that the thing is controlled using your cell phone or tablet computer. I would much, much prefer a solid, reliable two-thumb dedicated toggle controller with a more powerful radio transmitter, though something with at least a small 3" LCD feedback screen. I'd also like something where every element, helicopter and controller, was run off USB-rechargables. I don't think that's asking for too much these days; in fact, it should be getting just about standard in new technology. That we're still running to the drug store for Duracells in 2012 boggles my mind (or, more appropriately, having to haul the batteries out and stick them in a wall charger instead of connecting the thing to a port in the car while you're driving to and fro boggles the mind, ahem).
I'm also not enthusiastic about the apparently fixed camera position. Being able to point the camera left and right is arguably redundant since simply turning the drone will achieve that, but not being able to vary the camera in the vertical axis, to look up underneath something or, more fundamentally, down to the ground, is a big, big hole in the concept for me—particularly given that they had the genius idea of placing the rotors such that they'd be completely out of the way. As it stands, it would appear the only way to achieve a downward focus at all is to angle the drone downward, which necessarily means it's moving forward, and the deeper the angle, the faster it sweeps. Being able to hover and take pictures of an element of interest, and slowly guide the drone down to it, would for me be central to the idea. A third thumb toggle could accomplish this; presumably, once the drone was hovering steadily, your thumbs would be freed up from flying the drone to angle the view. I can't think of too many instances where changing the camera angle in flight would be crucial; if it is, there's probably some clever lad out there who's got the problem of too many controls not enough thumbs sussed already, I'm sure.
I was rather concerned about the thing flying out of range, especially vertically, but now that I've seen these videos and it's clear that you can set limits on how high the drone flies, it should simply be a matter of being reasonably close to what you want to photograph, and keeping an eye on the battery power. Still, three hundred bucks is a lot of money to have just vanish into the trees and rot till some kid finds it a decade hence.
What really got me thinking about all this was my expedition to the lost bridge on Given Road just about two years ago now. The bridge still remains, about fifty years after the last car used it, but its decking is long gone and I was too timid to attempt to cross it, especially since I was on my own. And I started musing about how great it would have been to have just had a little helicopter with a camera on it to just flying over, under, and around that bridge and get some interesting shots of it for three or four minutes. This is exactly the kind of thing that could achieve that.
I'm tempted to save my pennies and see about it in the spring, but the control means in particular is kind of a show-stopper for me, and the inability to tilt the camera without tilting the drone, so far as I can tell, takes a lot of the punch out of the idea. Maybe I should wait and see what version 3.0 offers. Anyway, it's at least now possible. I wonder what Ted Chirnside and James Salmon would have made of this.